Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320486502> ?p ?o ?g. }
- W4320486502 endingPage "113" @default.
- W4320486502 startingPage "104" @default.
- W4320486502 abstract "Abstract Recent years have witnessed a surge of interest in learning representations of graph-structured data, with applications from social networks to drug discovery. However, graph neural networks, the machine learning models for handling graph-structured data, face significant challenges when running on conventional digital hardware, including the slowdown of Moore’s law due to transistor scaling limits and the von Neumann bottleneck incurred by physically separated memory and processing units, as well as a high training cost. Here we present a hardware–software co-design to address these challenges, by designing an echo state graph neural network based on random resistive memory arrays, which are built from low-cost, nanoscale and stackable resistors for efficient in-memory computing. This approach leverages the intrinsic stochasticity of dielectric breakdown in resistive switching to implement random projections in hardware for an echo state network that effectively minimizes the training complexity thanks to its fixed and random weights. The system demonstrates state-of-the-art performance on both graph classification using the MUTAG and COLLAB datasets and node classification using the CORA dataset, achieving 2.16×, 35.42× and 40.37× improvements in energy efficiency for a projected random resistive memory-based hybrid analogue–digital system over a state-of-the-art graphics processing unit and 99.35%, 99.99% and 91.40% reductions of backward pass complexity compared with conventional graph learning. The results point to a promising direction for next-generation artificial intelligence systems for graph learning." @default.
- W4320486502 created "2023-02-14" @default.
- W4320486502 creator A5005469683 @default.
- W4320486502 creator A5008171524 @default.
- W4320486502 creator A5008931896 @default.
- W4320486502 creator A5011604061 @default.
- W4320486502 creator A5019056174 @default.
- W4320486502 creator A5049088847 @default.
- W4320486502 creator A5049715624 @default.
- W4320486502 creator A5049920459 @default.
- W4320486502 creator A5064842058 @default.
- W4320486502 creator A5068851680 @default.
- W4320486502 creator A5071605573 @default.
- W4320486502 creator A5077687075 @default.
- W4320486502 creator A5080087048 @default.
- W4320486502 creator A5083897792 @default.
- W4320486502 creator A5085028455 @default.
- W4320486502 creator A5090042508 @default.
- W4320486502 date "2023-02-13" @default.
- W4320486502 modified "2023-10-17" @default.
- W4320486502 title "Echo state graph neural networks with analogue random resistive memory arrays" @default.
- W4320486502 cites W1542981317 @default.
- W4320486502 cites W1975647260 @default.
- W4320486502 cites W1980490877 @default.
- W4320486502 cites W2004823737 @default.
- W4320486502 cites W2008857988 @default.
- W4320486502 cites W2056562706 @default.
- W4320486502 cites W2070856049 @default.
- W4320486502 cites W2071947829 @default.
- W4320486502 cites W2074357625 @default.
- W4320486502 cites W2116341502 @default.
- W4320486502 cites W2118706537 @default.
- W4320486502 cites W2139906443 @default.
- W4320486502 cites W2153513200 @default.
- W4320486502 cites W2155954834 @default.
- W4320486502 cites W2171865010 @default.
- W4320486502 cites W2585407525 @default.
- W4320486502 cites W2613205562 @default.
- W4320486502 cites W2762731122 @default.
- W4320486502 cites W2769049661 @default.
- W4320486502 cites W2772397789 @default.
- W4320486502 cites W2778935320 @default.
- W4320486502 cites W2782791387 @default.
- W4320486502 cites W2785141883 @default.
- W4320486502 cites W2788919350 @default.
- W4320486502 cites W2793600158 @default.
- W4320486502 cites W2800264288 @default.
- W4320486502 cites W2803163155 @default.
- W4320486502 cites W2805362231 @default.
- W4320486502 cites W2910318317 @default.
- W4320486502 cites W2913825337 @default.
- W4320486502 cites W2914721378 @default.
- W4320486502 cites W2948035163 @default.
- W4320486502 cites W2960778947 @default.
- W4320486502 cites W2962903741 @default.
- W4320486502 cites W2964051675 @default.
- W4320486502 cites W2964263004 @default.
- W4320486502 cites W2979589704 @default.
- W4320486502 cites W2997785591 @default.
- W4320486502 cites W2997992893 @default.
- W4320486502 cites W3003821665 @default.
- W4320486502 cites W3010339880 @default.
- W4320486502 cites W3013080934 @default.
- W4320486502 cites W3032819016 @default.
- W4320486502 cites W3035400263 @default.
- W4320486502 cites W3039560462 @default.
- W4320486502 cites W3042167479 @default.
- W4320486502 cites W3044192339 @default.
- W4320486502 cites W3046016807 @default.
- W4320486502 cites W3049461533 @default.
- W4320486502 cites W3080915835 @default.
- W4320486502 cites W3091977376 @default.
- W4320486502 cites W3098480967 @default.
- W4320486502 cites W3100848837 @default.
- W4320486502 cites W3101272433 @default.
- W4320486502 cites W3112668161 @default.
- W4320486502 cites W3122214732 @default.
- W4320486502 cites W3125908175 @default.
- W4320486502 cites W3171752851 @default.
- W4320486502 cites W3204457379 @default.
- W4320486502 cites W4254436426 @default.
- W4320486502 doi "https://doi.org/10.1038/s42256-023-00609-5" @default.
- W4320486502 hasPublicationYear "2023" @default.
- W4320486502 type Work @default.
- W4320486502 citedByCount "9" @default.
- W4320486502 countsByYear W43204865022023 @default.
- W4320486502 crossrefType "journal-article" @default.
- W4320486502 hasAuthorship W4320486502A5005469683 @default.
- W4320486502 hasAuthorship W4320486502A5008171524 @default.
- W4320486502 hasAuthorship W4320486502A5008931896 @default.
- W4320486502 hasAuthorship W4320486502A5011604061 @default.
- W4320486502 hasAuthorship W4320486502A5019056174 @default.
- W4320486502 hasAuthorship W4320486502A5049088847 @default.
- W4320486502 hasAuthorship W4320486502A5049715624 @default.
- W4320486502 hasAuthorship W4320486502A5049920459 @default.
- W4320486502 hasAuthorship W4320486502A5064842058 @default.
- W4320486502 hasAuthorship W4320486502A5068851680 @default.
- W4320486502 hasAuthorship W4320486502A5071605573 @default.